基于乘積偏好關系的專家模糊核聚類賦權方法
發(fā)布時間:2018-07-24 07:48
【摘要】:多屬性、多目標性決策中,針對專家給出各方案偏好關系下的決策問題,提出一種基于乘積偏好關系的專家模糊核聚類賦權方法。該方法運用模糊核聚類的思想實現(xiàn)對決策專家的聚類,并通過放寬歸一化約束條件,克服了傳統(tǒng)模糊核聚類算法中離群點對聚類結果的影響。同時,在專家類內賦權過程中,運用CI-IOWG算子集結同類專家的意見,依據不同專家對于形成類別一致性意見的貢獻程度來確定專家權重;克服了傳統(tǒng)基于熵權或判斷矩陣一致性的賦權方法的局限性。算例表明,該方法可行、有效。
[Abstract]:In multi-attribute and multi-objective decision making, an expert fuzzy kernel clustering method based on product preference relation is proposed to solve the problem of expert preference relationship. This method uses the idea of fuzzy kernel clustering to realize the clustering of decision experts, and by relaxing the normalized constraint conditions, it overcomes the influence of outliers on the clustering results in the traditional fuzzy kernel clustering algorithm. At the same time, in the process of experts' intra-class weighting, the expert weight is determined according to the contribution of different experts to the formation of category consistency by using CI-IOWG operator to gather the opinions of the same kind of experts. It overcomes the limitation of traditional weighting method based on entropy weight or consistency of judgment matrix. An example shows that this method is feasible and effective.
【作者單位】: 空軍工程大學裝備管理與安全工程學院;
【基金】:“十二五”國防預研基金資助項目(51327020104)
【分類號】:O225
本文編號:2140705
[Abstract]:In multi-attribute and multi-objective decision making, an expert fuzzy kernel clustering method based on product preference relation is proposed to solve the problem of expert preference relationship. This method uses the idea of fuzzy kernel clustering to realize the clustering of decision experts, and by relaxing the normalized constraint conditions, it overcomes the influence of outliers on the clustering results in the traditional fuzzy kernel clustering algorithm. At the same time, in the process of experts' intra-class weighting, the expert weight is determined according to the contribution of different experts to the formation of category consistency by using CI-IOWG operator to gather the opinions of the same kind of experts. It overcomes the limitation of traditional weighting method based on entropy weight or consistency of judgment matrix. An example shows that this method is feasible and effective.
【作者單位】: 空軍工程大學裝備管理與安全工程學院;
【基金】:“十二五”國防預研基金資助項目(51327020104)
【分類號】:O225
【相似文獻】
相關期刊論文 前10條
1 張國權;李文立;王明征;;基于離差函數和聯(lián)合熵的組合賦權方法[J];管理學報;2008年03期
2 李明奇;劉玉娟;趙美玲;;一種基于判斷矩陣的專家賦權方法[J];科技信息;2010年21期
3 席雪紅;;基于熵—Shapely的樣本差異賦權方法[J];統(tǒng)計與決策;2012年20期
4 許葉軍;達慶利;;不確定有序加權幾何平均算子的賦權方法(英文)[J];Journal of Southeast University(English Edition);2008年01期
5 申錦標;呂躍進;;一種基于向量貼近度的組合賦權方法[J];重慶工學院學報(自然科學版);2009年02期
6 張玉;魏華波;;基于CRITIC的多屬性決策組合賦權方法[J];統(tǒng)計與決策;2012年16期
7 陳瓊華;綜合評價中的賦權方法[J];統(tǒng)計與決策;2004年04期
8 楊超,李志強;基于主成分—神經網絡相集成的多策略評價賦權方法[J];當代財經;2005年07期
9 朱建軍;吳偉麗;劉思峰;;一種基于模糊線性規(guī)劃的主客觀組合賦權方法[J];運籌與管理;2006年03期
10 ;[J];;年期
相關會議論文 前1條
1 鄭凱;;對統(tǒng)計分析中常用賦權方法及權值效用的分析[A];第七屆全國體育科學大會論文摘要匯編(二)[C];2004年
,本文編號:2140705
本文鏈接:http://sikaile.net/kejilunwen/yysx/2140705.html